PROJECT ABSTRACT In the United States, adolescent engagement in substance use behaviors (SUBs) contributes to the leading causes of disease and death among youth (Eaton et al., 2012). Adolescence is a critical developmental period for the initiation and onset of substance use disorders that result in pervasive health consequences (Mahalik et al., 2013)., The economic costs associated with underage drinking have been estimated at $68 billion per year and an approximated $14 billion in juvenile justice costs (CASA, 2009). Taken together, adolescent SUBs are a serious public health concern with pernicious personal and societal consequences. Adolescent risky decision-making studies suggest that increases in SUBs stem from significant maturational changes in brain function resulting in adolescents' hypersensitivity to rewarding outcomes (e.g., Casey et al., 2008). Work in adults has shown that responses to uncertainty (i.e., probability of receiving such rewards) are equally as important in guiding decisions (e.g., Huettel et al., 2006). A type of uncertainty that may be pertinent to real- world risky behaviors are situations of ambiguity, in which information regarding the probabilities of outcomes is partially to completely unknown (e.g., Hsu et al., 2005). Here, we seek to investigate individual differences in adolescent responses to ambiguity when making decisions, and whether these differences relate to SUBs. An important factor in how adolescents respond to ambiguity may derive from subjective estimations of their chances of experiencing good or bad outcomes. Some studies have suggested that adolescents are biased in that they overestimate the likelihood of positive outcomes and underestimate negative ones, but research is limited (for review, see Reyna & Farley, 2006). Given that prevention efforts focus on educating adolescents on the consequences associated with risky behaviors, it is vital to understand how the adolescent brain perceives and updates beliefs about ambiguous outcomes. In addition, in order to create targeted methods by which unhealthy patterns of behavior can be prevented or changed, it is also important to understand the decision processes related to behaviors that protect against the adoption of SUBs. Engagement in protective behaviors (PBs) such as participation in structured team sports and extracurricular activities are related to positive health outcomes as well as decreased use of alcohol and other drugs (e.g.,Pate et al., 2000). To address these gaps in the literature, I propose the following approach. First, I plan to computationally model adolescent preferences of uncertainty during an economic lottery choice task and investigate their neural underpinnings using functional magnetic resonance imaging. Second, using a novel paradigm, I plan to measure adolescent biases and errors in estimations of uncertain outcomes and associated neural responses. In using this approach, I aim to characterize behavioral and biological markers for early identification of at-risk adolescents that may inform interventions that foster PBs over SUBs.